Infants at elevated likelihood for or later diagnosed with autism typically have smaller vocabularies than their peers, as shown by the Mullen Scales of Early Learning (MSEL) and the MacArthur-Bates Communicative Developmental Inventory (CDI). However, the extent to which MSEL and CDI scores align remains unclear, especially across clinical and non-clinical populations. This study examined whether the concurrent validity of the MSEL and CDI differs based on autism likelihood and diagnosis. Data from 720 14-month-old infants were analysed, grouped by likelihood (elevated vs. typical) and diagnosis at 36 months (diagnosed vs. not diagnosed). Vocabulary scores were compared across both likelihood and diagnostic groups. Moderate correlations were observed between the MSEL and CDI in most groups (rs range = [.34–.58]). One exception was that the expressive scores of elevated likelihood infants on the MSEL and CDI were more closely associated than the expressive scores of typical likelihood infants. Diagnosed infants had lower vocabulary scores than non-diagnosed peers on both the MSEL and CDI. The elevated likelihood group showed lower scores on the MSEL but not the CDI compared to typical likelihood infants. The moderate correlations suggest that the MSEL and CDI assess different aspects of language in infancy. These associations were weaker than previously reported in autistic children. Differences in vocabulary scores across likelihood and diagnostic groups highlight the need for further research to understand the association between these measures.